Posted on 17 August, 2015
The lab participated in the Dengue Forecasting Project, hosted by various federal government agencies.
The prediction challenge has two phases. All teams have submitted a first round of predictions, showing that they have a model that can make some predictions. Within the week, another round of data will be delivered to each team. At this time, the teams need to deliver “prospective” predictions on the new data.
Evan and Krzysztof spearheaded an effort to use state-space reconstruction to predict the course of dengue epidemics. You can check out (or fork) that work on GitHub.
For another team, Nick and Steve contributed weak learners to a random forest ensemble approach put together by Justin Lessler of the Johns Hopkins Infectious Disease Dynamics Working Group.
Which model will do better? Stay tuned…